Model comparison
DeepSeek-V3.1 vs Llama 2-70B
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 24.4 on the Noometry Index.
Last verified . 20 shared benchmarks.
DeepSeek
MetaSummary
- They share 20 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 8 categories and Llama 2-70B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-V3.1 leads 43.7 to 7.4.
- The biggest single-benchmark swing is DTBench: 82.7% for DeepSeek-V3.1 and 41.6% for Llama 2-70B.
Side by side
| DeepSeek-V3.1 | Llama 2-70B | |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 42.8 | 24.4 |
| Released | 2025-08-21 | 2023-07-18 |
| Weights | Open | Open |
| Context window | 164K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $0.95 | — |
| Results tracked | 27 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1 leads
DeepSeek-V3.1: 40.3 (#144), Llama 2-70B: 31.4 (#286)
| Benchmark | DeepSeek-V3.1 | Llama 2-70B |
|---|---|---|
| LMArena Coding | 1417 | 1079 |
| WeirdML | 38.4% | — |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.9 (#110), Llama 2-70B: 14.4 (#325)
| Benchmark | DeepSeek-V3.1 | Llama 2-70B |
|---|---|---|
| LMArena Hard Prompts | 1417 | 1073 |
| DTBench | 82.7% | 41.6% |
| Epoch Capabilities Index | 139.92 | 113.79 |
| ForecastBench | 58 | 51.4 |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| LMCA | 24.3% | — |
| BIG-Bench Hard | — | 64.9% |
| CommonsenseQA 2.0 | — | 50% |
| HellaSwag | — | 85.3% |
| LAMBADA | — | 78.9% |
| PIQA | — | 82.8% |
| WinoGrande | — | 80.2% |
Math DeepSeek-V3.1 leads
DeepSeek-V3.1: 38.9 (#122), Llama 2-70B: 8.1 (#326)
| Benchmark | DeepSeek-V3.1 | Llama 2-70B |
|---|---|---|
| LMArena Math | 1420 | 1091 |
| OTIS Mock AIME 2024-2025 | — | 0% |
| MATH Level 5 | — | 3.3% |
| GSM8K | — | 69.6% |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.7 (#90), Llama 2-70B: 7.4 (#310)
| Benchmark | DeepSeek-V3.1 | Llama 2-70B |
|---|---|---|
| LMArena Expert | 1405 | 1039 |
| GPQA Diamond | — | 26.3% |
| Vectara Hallucination Rate | 5.5% | — |
| ARC (AI2) Challenge | — | 78.3% |
| BoolQ | — | 88.6% |
| MMLU | — | 69.9% |
| OpenBookQA | — | 60.2% |
| TriviaQA | — | 87.6% |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.6 (#106), Llama 2-70B: 27.7 (#274)
| Benchmark | DeepSeek-V3.1 | Llama 2-70B |
|---|---|---|
| LMArena Non-English | 1400 | 1045 |
| LMArena Chinese | 1469 | 995 |
| LMArena French | 1447 | 1090 |
| LMArena German | 1411 | 1041 |
| LMArena Japanese | 1378 | 927 |
| LMArena Korean | 1337 | 964 |
| LMArena Russian | 1405 | 1083 |
| LMArena Spanish | 1431 | 1143 |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-V3.1: 73.9 (#110), Llama 2-70B: 54.9 (#278)
| Benchmark | DeepSeek-V3.1 | Llama 2-70B |
|---|---|---|
| LMArena Instruction Following | 1400 | 1071 |
Long Context DeepSeek-V3.1 leads
DeepSeek-V3.1: 36.3 (#232), Llama 2-70B: 32.3 (#270)
| Benchmark | DeepSeek-V3.1 | Llama 2-70B |
|---|---|---|
| LMArena Longer Query | 1422 | 1062 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.3 (#98), Llama 2-70B: 32.3 (#279)
| Benchmark | DeepSeek-V3.1 | Llama 2-70B |
|---|---|---|
| LMArena Text | 1420 | 1115 |
| LMArena Creative Writing | 1401 | 1075 |
| LMArena Multi-Turn | 1408 | 1088 |
| EQ-Bench Creative Writing | 1436 | — |
Frequently asked questions
Is DeepSeek-V3.1 better than Llama 2-70B?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 24.4 on the Noometry Index.
Is DeepSeek-V3.1 or Llama 2-70B better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 31.4 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1 and Llama 2-70B share?
20 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Llama 2-70B has 35.